Loading field map
Is the meaning of a sentence determined by how it relates to the world—by its truth conditions and reference—or by how it is used in communication—by the intentions, inferences, and contexts of speakers and hearers? This tension between world-directed and practice-directed accounts has driven the philosophy of language for centuries. The subfield of meaning and understanding is the story of competing frameworks that have offered different answers, each emerging from the perceived shortcomings of its predecessors.
The first systematic framework in the Western tradition was the Scholastic Philosophy of Language (1100–1500). Medieval logicians such as Peter Abelard and William of Ockham developed theories of signification and supposition. For the Scholastics, a term signifies a concept in the mind, and that concept stands for (supposits for) things in the world. Meaning was thus a three-way relation between words, concepts, and reality. This framework was primarily concerned with the logic of terms and propositions, and it treated meaning as a stable property of linguistic expressions. It would later be revived in part by inferentialists, but for centuries it was the dominant approach.
By the late nineteenth century, dissatisfaction with psychologistic and historical approaches to meaning led to a new emphasis on logical form. Ideal Language Philosophy (1880–1930), associated with Gottlob Frege, Bertrand Russell, and the early Ludwig Wittgenstein, held that natural language is logically defective and must be replaced or regimented into a logically perfect notation. Frege’s distinction between sense and reference showed that meaning involves both a mode of presentation (sense) and an object (reference). This framework treated meaning as a matter of truth conditions and compositionality: the meaning of a complex expression is a function of the meanings of its parts.
Wittgenstein’s Picture Theory of Meaning (1920–1930), presented in the Tractatus Logico-Philosophicus, radicalized this idea: a proposition is a picture of reality, and its meaning is the state of affairs it depicts. The picture theory shared with Ideal Language Philosophy the assumption that meaning is a relation between language and the world, but it added a metaphysical claim about the structure of representation.
Logical Positivism (1920–1950) took the logical turn further by adding a verificationist criterion of meaning. For the Vienna Circle, a statement is meaningful only if it is either analytically true or empirically verifiable. This narrowed the scope of meaningful discourse to science and logic, dismissing metaphysics, ethics, and aesthetics as cognitively meaningless. Logical Positivism thus absorbed the picture theory’s focus on factual content but replaced picturing with verification as the key to meaning.
The logical turn’s disregard for ordinary language provoked a strong reaction. Ordinary Language Philosophy (1930–1970), led by the later Wittgenstein, J. L. Austin, and Gilbert Ryle, argued that philosophical problems arise from misusing everyday language. Wittgenstein’s slogan “meaning is use” encapsulated a radical shift: instead of looking for a single relation between words and the world, philosophers should examine the diverse ways language functions in concrete social practices. This framework rejected the Ideal Language assumption that ordinary language is defective and instead treated it as the ultimate arbiter of philosophical questions.
Gricean Pragmatics (1950–1970) offered a more systematic account of use. H. P. Grice distinguished between sentence meaning and speaker meaning, arguing that what a speaker means is a matter of their intentions. He introduced the cooperative principle and conversational implicatures to explain how hearers infer meaning beyond what is literally said. Gricean pragmatics coexisted with truth-conditional semantics by treating speaker meaning as an additional layer, not a replacement.
Speech Act Theory (1950–1970), developed by Austin and John Searle, treated utterances as actions. Austin’s performative utterances (e.g., “I promise”) do not describe but do things; his later theory of illocutionary acts classified the forces utterances can have (asserting, questioning, ordering). Speech Act Theory absorbed Grice’s emphasis on intention but added a social dimension: meaning depends on conventional rules and institutional contexts. It remains a living tradition, especially in feminist and institutional speech act theory.
While the pragmatic revolution flourished, a new wave of formal approaches reasserted the centrality of truth and reference. Truth-Conditional Semantics (1960–1990), inspired by Donald Davidson and Alfred Tarski, held that to give the meaning of a sentence is to state its truth conditions. Davidson argued that a compositional theory of meaning can be built on a recursive definition of truth. This framework revived the Ideal Language project but without the hostility to natural language: it aimed to model natural language semantics using formal tools.
Causal Theory of Reference (1970–1990), advanced by Saul Kripke and Hilary Putnam, challenged the descriptivist assumption that names and natural kind terms have their meaning fixed by descriptions. Instead, reference is determined by an initial baptism and a causal chain of use. This sparked the internalism–externalism debate: whether meaning is in the head (internalist) or depends on the environment (externalist). The Causal Theory coexisted with Truth-Conditional Semantics but narrowed its focus to reference, leaving other aspects of meaning to pragmatics.
Formal Semantics (1970–Present), pioneered by Richard Montague, applied model-theoretic methods to natural language. Montague grammar treated English as a formal language amenable to mathematical analysis. Formal Semantics absorbed Truth-Conditional Semantics and extended it to a wide range of phenomena (quantification, tense, modality). It remains the dominant framework in linguistics, providing precise compositional analyses. Its success has led to a division of labor: Formal Semantics handles sentence meaning, while pragmatics handles context-dependent aspects.
Since the 1980s, several frameworks have challenged the static, truth-conditional picture. Dynamic Semantics (1980–Present) reinterprets meaning as context change potential. Instead of a sentence having a truth condition, it has the capacity to update the common ground of a conversation. This framework, developed by Jeroen Groenendijk, Martin Stokhof, and others, handles anaphora and presupposition more naturally than static semantics. Dynamic Semantics coexists with Formal Semantics, often as a refinement rather than a replacement.
Inferentialism (1980–Present), championed by Robert Brandom, rejects the representationalist assumption that meaning is a relation between language and the world. Instead, meaning is constituted by inferential roles: to understand an expression is to master the inferences it licenses. Inferentialism revives elements of Scholastic supposition theory and Ordinary Language Philosophy’s emphasis on use, but it offers a systematic normative account. It stands in living disagreement with truth-conditional approaches, arguing that truth conditions are derivative from inferential practices.
Contextualism (1990–Present) holds that the truth conditions of an utterance are heavily dependent on context, far beyond what indexicals and demonstratives require. Proponents like François Recanati and Dan Sperber argue that pragmatic processes (e.g., free enrichment, ad hoc concept construction) shape what is said. Contextualism narrows the scope of semantic minimalism, which holds that sentence meaning is relatively stable. The debate between contextualists and semantic minimalists is one of the most active today.
Relativism about Truth (2000–Present) extends contextualist insights by arguing that the truth of certain utterances (e.g., about taste, knowledge, or future contingents) is relative to a context of assessment. John MacFarlane and others have developed frameworks where truth is not absolute but varies with the assessor’s perspective. Relativism about Truth coexists with Contextualism but goes further, challenging the very idea of a single truth value for a proposition.
Today, the leading frameworks are Formal Semantics, Dynamic Semantics, Inferentialism, Contextualism, and Relativism about Truth. They agree that meaning is systematic and that context matters, but they disagree on what the fundamental unit of meaning is. Formal Semantics and Dynamic Semantics treat truth conditions or context change as central; Inferentialism treats inferential role as primary; Contextualism and Relativism emphasize the pervasive role of context and assessment. The internalism–externalism debate remains unresolved, with externalist causal theories still influential. The division of labor between semantics and pragmatics is constantly renegotiated. The subfield’s history shows that no single framework has won: each captures important aspects of meaning, and the tension between world-directed and practice-directed accounts continues to drive inquiry.